Pain, ( 3 1 International Associationfor the Study of Pain. All rights reserved0304-3959/96/$15.00 PAIN 2992 A parsimonious set of verbal descriptors of pain sensation derived from the McGill Pain Questionnaire Ephrem Fernandez* and Stuart Towery D P S M U D T 7 ( (Received8 June 1995;revisionreceived22 August 1995;accepted20 September1995) s In viewof some recent disagreementsabout the vocabulary of pain as suggestedin the McGill Pain Questionnaire(MPQ), the present study re-examinedall MPQ pain descriptorswith regard to their appropriateness as descriptorsof pain sensation.A sampleof 70 undergraduatestudents (whosefirst languagewas English)assigned descriptorsto individualsensorysubcategoriesand then rated them in terms of impliedintensityof pain. Data were evaluated using three criteria related to the absolute frequency, relative frequency, and unimodality of word assignmentsto each subcategory.Results revealedthat about 400/0of the supposedMPQ sensorydescriptorscould not beclassifiedwithinany ofthe sensorysubcategoriesbecauseofincomprehension,underuse,or ambiguityof usage. The majority of the words, however,were classifiedin the same subcategoriesand given similarintensityratings as in the MPQ. Thesewordsconstitutea parsimonioussubsetof MPQ descriptorsof pain sensation.Suchwords promise more diagnosticspecificityin the assessmentof pain. Further research could serveto replicatethese findingsas part of the ongoing refinementof the MPQ. Pain descriptors;Sensation;Classification;MPQ; Decisionrule I Becausepain is a subjective,private experience,lan- guage offers a useful method of labelingthis phenome- non; apart from its value in mental representation, languagealso servesas a tool in the communicationof pain to others. Words may be used to convey multidimensionalproperties of pain, such as intensity, sensation,and affectivereaction,whichare of particular diagnosticvalue to clinicians. The first systematic attempt to incorporate verbal descriptors in pain assessment was the McGill Pain Questionnaire or MPQ (Melzack 1975),currently the most widelyused instrumentin pain research and prac- tice. The MPQ is a self-reportinstrument that presents 78 pain dewriptors distributed across 20 subcategories (with an additional 6 words in the Present Pain Index); these are further organizedinto thrm broad sensory, *C 7 7 e-mail:efemand@mail.cis. sun.edu. ‘ p T S S 0 3 and categoriesas wellas an additional m category.The sensorycategoryencompas- ses10subcategories: p c b and For example,the word contained within the subcategory whereas the word falls within the sub- catego~. The implicationis that each word reflects a particular sensoryproperty of pain. The affectivecate- gory containswords such as and which are supposed to reflect negativeemotion and its autonomic properties. Finally, the evaluativecategory containswordsrangingfrom to e which reflect the overall intensity of pain. In addition, each word from each categoryhas a rank value indicativeof the relativeintensityof pain implied. Psychometrically,the MPQ meets severalimportant criteria. Adequate test-retest reliability has been demonstratedfor someof the word subcategories(Love et al. 1989)and internal consistencyhas been shown in the case of cancer patients (Graham et al. 1980).MPQ ratingshave also been found to be sensitiveto standar- dized stimuli (Klepac et al. 1981), and capable of discriminatingbetweenacute and chronicpain (Teman